A global registry (COV-GN) was set up to collect clinical data of patients with COVID-19 and established or newly diagnosed glomerular disease, which presented to several nephrology departments worldwide during the early (pre-omicron) pandemic era from March 2020 until August 2021.

The criteria for inclusion in this study were i) an established diagnosis of COVID-19, either verified via viral RNA detection or because of high clinical probability, e.g. by typical symptoms and contact to a verified case and/or characteristic radiologic findings on CT scan; ii) age ≥ 18 years at the time of COVID-19, and iii) a previously established diagnosis of a immune-mediated glomerular disease. Secondary forms of glomerular diseases and glomerular diseases that clearly are not immune-mediated (e.g. diabetic kidney disease) were excluded.

Participating centers were recruited via announcements by the European Renal Association- Immunonephrology Working Group (ERA-IWG) or approached directly.

Study data were collected and managed using REDCap (Research Electronic Data Capture) electronic data capture tools hosted at Medical University Innsbruck (13, 14). REDCap is a secure, web-based software platform designed to support data capture for research studies. An electronic case report form was established and transformed onto REDCap using a web-integrated project designer. Participating centers received access to an individual REDCap account. Sub-investigators screened local patient data for eligibility and entered pseudonymized patient records assigned to their respective center via their REDCap accounts. The online survey was launched on 29.06.2020 and closed on 31.03.2022.

The following data were obtained: sex, age at diagnosis of COVID-19, height, weight, race, diagnosis of established glomerular disease, remission status (complete or partial remission, no remission, relapse) and immunosuppressive regimen (induction treatment, maintenance treatment, no immunosuppression) at time of COVID-19, clinical data on admission including O2 saturation and body temperature. Laboratory data were collected on admission, peak values during follow-up and one month (3 to 5 weeks) after diagnosis of COVID-19 and included urea, creatinine, proteinuria, albuminuria, C-reactive protein, serum albumin. Clinical outcomes of COVID-19 were assessed in categories (hospitalization, non-invasive ventilation, intensive care treatment, mechanical ventilation, kidney replacement therapy due to AKI, and mortality). Further information on prior treatment of glomerular disease included name of the medication, cumulative dose, treatment duration, and the time interval from last application of immunosuppressants used.

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of the Medical University Innsbruck (1215/2020).

The clinical outcome of COVID-19 was assessed in categories. In the survey, multiple answers could be selected. For statistical analysis, outcomes were ranked by severity, increasing from lowest to highest severity. Outcomes were further classified into two groups. Patients that were managed without hospitalization (non-severe COVID-19) were compared with those that were hospitalized, treated with non-invasive ventilation, intensive care unit (ICU) care, mechanical ventilation or died (severe COVID-19), according to the most severe outcome selected. Remission status of established glomerular disease (complete remission, partial remission, no response, and relapse) as judged by the investigator on the last clinical visit before COVID-19 was assessed and dichotomized into a “remission” group (complete + partial remission) and a “no remission” group (no response + relapse). Immunosuppressive regimen at the time of COVID-19 was assessed in three categories (induction treatment, maintenance treatment, no immunosuppression) and dichotomized into an “ongoing immunosuppression” group (induction treatment + maintenance treatment) and a “no ongoing immunosuppression” group (no immunosuppression). AKI was defined as recommended by the Acute Disease Quality Initiative (ADQI) Workgroup (15), according to KDIGO criteria, as Δ serum creatinine ≥ 0.3 mg/dl from the time of admission to peak follow-up value or new start of kidney replacement treatment due to AKI. Furthermore, AKI was staged for severity according to KDIGO criteria with peak serum creatinine 1.5-1.9 times baseline for stage 1, 2.0-2.9 times baseline for stage 2 and ≥ 3.0 times baseline or increase in serum creatinine to ≥ 4.0 mg/dL or initiation of kidney replacement therapy for stage 3 (16). To test the differences between measures over time, paired t-tests were considered. Because of the rather low sample sizes, regression analyses were not used. For differences between groups, t-tests were calculated. Mann-Whitney-U-tests were used to test the differences between dichotomous variables.

The study included 59 patients from 11 centers, covering Europe, North and South America and Asia. Patients had a previously established diagnosis of an immune-mediated glomerular disease and were diagnosed with COVID-19 between 01.03.2020 and 31.08.2021. Baseline characteristics are illustrated in Table 1 .

The most common diagnoses were lupus nephritis (32.2%), followed by anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis (18.6%), focal segmental glomerulosclerosis (FSGS) (16.9%), IgA nephropathy (13.6%) and membranous nephropathy (8.5%). About two thirds (61%) of included patients were in complete or partial remission before COVID-19, while about one third (35.6%) had insufficient glomerular disease control; 55.9% were on immunosuppressive treatment (either induction or maintenance regimen) and 40.7% were without immunosuppressive treatment at presentation.

Mean serum creatinine and mean proteinuria at presentation were 2.76 ± 3.39 mg/dL and 2.60 ± 4.93 g/g, as assessed by spot urine measurement (urinary protein/creatinine ratio) or 24-hour urine collection (g/24h), respectively. During follow-up, serum creatinine increased to a peak mean of 3.28 ± 4.0 mg/dL (mean_{difference to baseline} = 0.22 ± 1.03, p = .227), while proteinuria decreased constantly during follow-up to a minimum of 2.07 ± 3.15 mg/g (mean_{difference to baseline} = -0.59 ± 3.31, p = .411). Both serum creatinine and proteinuria recovered one month after admission to a mean of 2.13 ± 1.72 mg/dL (mean_{difference to baseline} = 0.81 ± 3.41, p = .176) and 2.07 ± 3.15 g/g or g/24h (mean_{difference to baseline} = -0.10 ± 2.04, p = .124), respectively. Overall, ten patients (16.9%) developed AKI, of whom two patients were classified as stage I and eight patients as stage III AKI. Among those with stage III AKI, four required intermittent hemodialysis, while no continuous kidney replacement therapy or peritoneal dialysis was performed. The course of kidney-specific parameters over time is illustrated in Figure 1 .

The clinical outcomes of COVID-19 ranked by severity are summarized in Table 2 . 44.1% of patients were managed in an outpatient setting. The remaining 55.9% were hospitalized (35.6%) or even required more intensive treatment modalities, including non-invasive ventilation (3.4%), intensive care treatment (6.8%), and mechanical ventilation (5.1%). Three patients (5.1%), two of whom were hospitalized, and one outpatient died due to respiratory failure.

AKI in the context of COVID-19 is associated with high mortality and is an established risk factor for in-hospital death in patients with COVID-19 (15). The exact underlying pathophysiology of COVID-19 associated AKI is not fully understood but several mechanisms are being discussed to be involved (18). Epidemiological data on the incidence of COVID-19 associated AKI is extremely variable and depending on several factors including geographic and temporal variations during the pandemic (19), but also the lack of use of common diagnostic criteria was associated with an underestimation of AKI rates (20). Accordingly, recommendations for the diagnosis, prevention and management of COVID-19 associated AKI have been published, e.g. by the Acute Disease Quality Initiative (ADQI) Workgroup (15).

Previously, only the IRoc-GN registry reported outcomes of patients with COVID-19 and glomerular disease in comparison to patients without glomerular disease.

The AKI rate in COV-GN was substantially lower than in patients with glomerular disease reported in the IRoc-GN registry (16.9% versus 39%) and comparable with those without glomerular disease included in IRoc-GN (16.9% versus 14%) (10). This difference may be explained by the difference in reporting baseline serum creatinine values. In COV-GN, development of AKI was measured from the time of COVID-19 diagnosis, while IRoc-GN assessed baseline kidney function at a median of 3.7 (interquartile range 2.3-9.6) months before infection (11). According to a meta-analysis, the overall prevalence of AKI in patients with COVID-19 during a comparable period of the pandemic was 17% and presence of AKI was significantly associated with severe COVID-19 (21). Likewise, the mortality rate of 5.1% in COV-GN was substantially lower than a 15% mortality rate reported in IRoc-GN (10). It needs to be emphasized that varying outcomes across studies are common and may be attributed to factors such as differences in patient characteristics, methodological aspects and different biases such as selection bias (22). IRoc-GN recruited mainly patients affected during the first wave of the pandemic, which involved limited testing capacities. Thus this cohort might have been prone to more severe disease outcomes. Moreover, differing baseline characteristics of glomerular disease patients included in IRoc-GN and COV-GN need to be considered (10). Patients in IRoc-GN were older (60.3 ± 17.7 years versus 47.48 ± 17.9 years, respectively) and included fewer male patients (42.5% versus 52.5%, respectively). Also, the proportion of hospitalized patients was higher in the IRoc-GN cohort (70%), as compared to COV-GN (54%). The composition of included disease entities was broadly comparable.

Lower serum albumin levels at presentation were associated with severe COVID-19 outcomes in this cohort. This finding is in accordance with previous reports, indicating that hypoalbuminemia predicts severe COVID-19 outcomes in the general population (23, 24). Similarly, Waldman et al. found that serum albumin at presentation was significantly lower in patients with AKI, those requiring kidney replacement therapy and those with venous thromboembolic events, but was not associated with increased mortality in the IRoc-GN cohort (10). Other COVID-19 related outcomes were not further analyzed in IRoc-GN. In COV-GN, however, risk factors for severe COVID-19 outcomes were analyzed by stratification into two clinically relevant and similarly large groups (severe versus non-severe). This may have advantages over mortality rate as sole marker for severe COVID-19 outcomes to identify high-risk patients requiring a particularly vigilant management.

Also, in accordance with results from IRoc-GN, immunosuppressive treatment at the time of COVID-19 was not associated with adverse outcomes in both cohorts. However, and in contrast to IRoc-GN, further analysis of immunosuppressant subgroups revealed ongoing intake of corticosteroids at presentation to be associated with severe COVID-19 outcomes. These observations go in line with results from a large retrospective population-based analysis of centralized health care data from England (NHS Digital) during the second wave of COVID-19 (25). Patients with rare autoimmune rheumatic disease showed a 2.76 times risk of COVID-19-related death than the general population, with corticosteroids associated with increased risk. Overall, the discussion about risks and benefits of immunosuppressants in the context of COVID-19 is conflicting (12). Results from a large retrospective registry of patients with different rheumatic diseases showed that both, higher disease activity but also the use of rituximab and daily prednisone doses ≥ 10 mg, were associated with COVID-19-related death (26). These findings might reflect that a recent higher disease activity necessitating higher doses of immunosuppressants portends a higher risk of severe COVID-19 outcomes, as would be expected. This was also the case in a Swedish report on ANCA-vasculitis and COVID-19, which found that COVID-19 was more common in patients with higher doses of prednisone. Significant risk factors for severe COVID-19 were impaired kidney function and induction therapy whereas maintenance rituximab was not associated with severe COVID-19 disease (27).

Male sex and older age were both significantly associated with severe COVID-19 in COV-GN. Both are established risk factors for severe COVID-19 outcomes, as confirmed repeatedly throughout the pandemic (26, 28–31). Recently, sex differences in COVID-19 mortality risk have been compared among patients on kidney replacement therapy (32). In a large cohort of 3206 dialysis patients and 1204 kidney transplant recipients, male sex was a risk factor for mortality in dialysis patients but not in kidney transplant recipients. The authors assume that the use of immunosuppressants in the latter group may have narrowed sex differences and argue that the otherwise lower risk of females may be attributed to a more robust immune response to COVID-19 compared with men (33). In this cohort, immunosuppressive treatment at the time of COVID-19 did not equalize sex differences. In accordance with this, male sex was an independent risk factor for COVID-19 related death in a large cohort of 3729 patients with different rheumatic diseases, of whom 88% (3290/3729) were on immunosuppressive or immunomodulatory treatment (26).

Apparent limitations of this study are its retrospective design, the relatively small cohort size, missing data at different time points and most importantly, the lack of a control group. A matched control group of patients with COVID-19 not affected by glomerular diseases would be beneficial in interpreting the results. However, even a perfectly matched control group would harbor significant risks in a retrospective study design. As an example, the indication for hospitalization will be based on other criteria in immunocompromised patients when compared with otherwise healthy individuals. Comparison with hospitalized patients only, as in IRoc-GN, provides limited advantage, particularly for the chosen outcome (severe versus non-severe COVID-19). Still, in absence of a control group, no firm conclusions on the risk of patients with glomerular diseases compared with the general population or patients with other forms of CKD is possible. Glomerular diseases are rare diseases and the establishment of the registry aimed to collect data on early evolving SARS-CoV2 strains in order to identify risk factors for severe disease courses. To recruit as many patients as possible, multiple centers worldwide were involved and inclusion criteria were defined broadly. Inclusion of patients at later time points of the pandemic would have increased the sample size, however, mass vaccination and less virulent strains would have created more heterogeneity and difficulties in interpretation of data. Although an effort was made to recruit patients globally, a relatively high proportion of patients with Caucasian ethnicity is limiting generalizability. Hospitalization of patients occurred upon discretion of the treating physician. Thus, the respective indications for hospitalization are not standardized and may vary in our cohort.

Vaccination status and currently available SARS-CoV-2-specific treatment options were initially not considered in our study. Considering the rollout of the first SARS-CoV-2 vaccines and recruitment from this time period in the respective countries, only a total of 5 patients were possibly exposed to one or more vaccination doses before presenting with COVID-19. Of these, only 1 patient received a single dose of the mRNA vaccine Spikevax^® (Moderna) only a few days before being diagnosed with COVID-19. No patient had a complete primary vaccination series (= fully vaccinated). Thus, the effect of SARS-CoV-2 vaccination can be neglected in the present cohort.

Our data suggest that during the pre-omicron period of the COVID-19 pandemic, patients with glomerular disease were frequently experiencing severe COVID-19 related outcomes. However, rates of severe outcomes including AKI, hospitalization and mortality were lower than in a previously published cohort of patients with glomerular disease, which may be attributed to different baseline characteristics of these cohorts and the inclusion of patients after the first wave of the pandemic, as prognosis in general improved. Of particular importance, higher age, male sex, ongoing intake of corticosteroids and lower serum albumin levels at presentation appear to be risk factors for severe COVID-19 in patients with glomerular diseases.